Genes conferring drug resistance can potentially be applied clinically in two different ways: (a) To confer drug-resistance upon normal, sensitive tissues, thereby averting toxic side-effects and allowing improved anti tumor chemotherapy; (h)To allow selective, in vivo expansion of cells engineered to express the drug resistance gene, perhaps along with other therapeutic genes. In this competing renewal application, experiments are described which will continue to address the use of methotrexate (MTX) as an anti-tumor or selective agent in combination with variant forms of drug- resistant dihydrofolate reductase (DHFR). There are three Specific Aims. In Aim 1, the mechanism by which drug-resistant DHFR gene expression may be used for improved chemotherapy will be addressed, by characterizing lympho-hematopoietic cell types which repopulate the intestine after transplantation of DHFR transgenic marrow and which may contribute to MTX resistance in the GI tract. These studies will also contribute to a greater understanding of the role of lymphohematopoietic cells in the maintenance of GI structure and function, especially in patients undergoing BMT and/or chemotherapy. In Aim 2, experiments will be conducted to determine the extent to which drug resistance conferred by transplantation with MTX-resistant DHFR transgenic marrow will allow improved survival of animals bearing the well-characterized L1210 and P388 model leukemias by administration of higher doses of MTX. MTX- mediated selective enrichment of drug-resistant DHFR expressing marrow will be tested directly in Aim 3 through competitive marrow recovery experiments in which animals transplanted with mixtures of normal and DHFR transgenic marrow will be allowed to engraft, and then administered combinations of antifolate (MTX, trimetrexate) with nucleoside transport inhibitor to prevent nucleic acid precursor salvage and circumvention of antifolate toxicity. Dosing regimens will be optimized, and the extent to which low levels of engrafted drug-resistant stem cells can be expanded will be determined. Results from these studies will address the feasibility of applying MTX resistance gene transfer for improved chemotherapy in humans as well as using MTX as an in vivo selective agent to deal with the low levels of gene transfer currently observed in humans and in large animals models, thus contributing a crucial preclinical component to the ongoing formulation of DHFR gene transfer clinical trials at the University of Minnesota.